Quality control for beer and other beverages may soon be electronic.
The matter of taste and how it is influenced by external factors is a subject of fascination for many people (witness the proliferation of polls about which sense you would give up if needed, for example). In the foodservice industry, the question of how people perceive flavors is big business, used to predict upcoming food trends and what will resonate with tomorrow’s fickle diner.
In a nice touch of irony, this inherently human sensory experience is being increasingly monitored — and replicated — by artificial intelligence and other technological advances.
What we register as taste often comes from the molecular composition of a food ingredient. The IBM technology uses combinatorial sensing to build a cross section of a liquid, creating a “holistic signal” using an array of sensors. “Hypertaste uses electrochemical sensors comprised of pairs of electrodes, each responding to the presence of a combination of molecules by means of a voltage signal, which is easy to measure. The combined voltage signals of all pairs of electrodes represents the liquid’s fingerprint,” writes Patrick Ruch on the IBM Research Blog. “Key to the functioning of our electrochemical sensors are polymer coatings covering each electrode. At our lab in Zurich, we synthesize these coatings which are designed to capture a range of chemical information and allow a high degree of miniaturization.”
An interesting note is that the sensor needs to be trained by exposure to the liquid multiple times, much like a human being trains their palate. Similar to human learning, the machines can also compare notes and exchange information to gain additional knowledge from a community of sensors in the field. Unlike human capabilities, however, the machine’s palate can be rewired by resetting the learning parameters remotely through the cloud.
The potential for taking these types of tech past the world of safety and into the more ephemeral idea of “good” tastes and “bad” ones is intriguing. Researchers in Spain have already used an electronic tongue to classify different types of beer. Using Dropsens screen printed electrodes, the machine uses four electrodes made of different material to determine the polyphenol concentration and differentiate between four styles of beers, according to Inside Science: alcohol-free, Pilsners, dopplebock and European strong lager. The electronic tongue could place all of the 25 types of beer within one of those categories with 100 per cent accuracy, its color with 76 per cent accuracy and its alcohol content with 84 per cent accuracy.
Sommeliers and cicerones don’t have to welcome their robot overlords — not yet, anyway. “Humans are still better,” María Luz Rodríguez-Méndez, a professor of inorganic chemistry at University of Valladolid in Spain and co-author of the study told Inside Science. “At least for now, they are far from being replaced.”